- คุณจุฑารัตน์ คีรีเพ็ชร นักวิจัย ทีมวิจัยคลังอนุพันธ์ความรู้ (KEA) กลุ่มวิจัยวิทยาการข้อมูลและการวิเคราะห์ (DSARG)
- ซึ่งผลงานได้รับการคัดเลือกให้เป็น 1 ใน 5 ผลงานเด่น คว้ารางวัลประจำปี 2020
- “Alteryx Analytics 2020 Excellence Award Winners”
- ในงานประชุมประจำปี Inspire APAC 2020
- – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –
- จากกรณีศึกษา (use case) การพัฒนา workflow เพื่อการจัดการและวิเคราะห์ข้อมูล geospatial ด้านการเกษตรในประเทศไทย ภายใต้โครงการ Agri-Map (What2Grow) หัวข้อ “𝗨𝘀𝗶𝗻𝗴 𝗔𝗹𝘁𝗲𝗿𝘆𝘅 𝘁𝗼 𝗗𝗲𝘃𝗲𝗹𝗼𝗽 𝗦𝗺𝗮𝗿𝘁 𝗙𝗮𝗿𝗺𝗶𝗻𝗴 𝗶𝗻 𝗧𝗵𝗮𝗶𝗹𝗮𝗻𝗱”
- ในงานประชุมประจำปี Inspire APAC 2020 ซึ่งจัดขึ้นระหว่างวันที่ 25-26 กุมภาพันธ์ที่ผ่านมา ณ นครซิดนีย์ ประเทศออสเตรเลีย โดยงานดังกล่าวเป็นการประชุมของกลุ่ม data scientist, analyst, IT & business leader ทั่วโลกผู้พัฒนาและสนับสนุนงานด้าน analysis work
Using Alteryx to Develop Smart Farming in Thailand
- Author: Jutarat Khiripet
𝗢𝘃𝗲𝗿𝘃𝗶𝗲𝘄 𝗼𝗳 𝗨𝘀𝗲 𝗖𝗮𝘀𝗲
Thailand’s National Electronics and Computer Technology Center (NECTEC) is a statutory government organization under the National Science and Technology Development Agency (NSTDA), Ministry of Higher Education, Science, Research and Innovation. Jutarat Khiripet, a researcher in Knowledge Elicitation and Archiving Laboratory (KEA) under NECTEC uses Alteryx to develop geospatial data analysis for agriculture land use management – in order to build the Smart Farming system in Thailand. By leveraging Alteryx, she will be able to save a substantial amount of man-hours so that her team can contribute to developing more meaningful work rather than spending days on manual tasks. As the operations are designed and developed to be an automated process workflow, it is going to help the team to manage the processes better in the long run (or moving forward).
𝗗𝗲𝘀𝗰𝗿𝗶𝗯𝗲 𝘁𝗵𝗲 𝗯𝘂𝘀𝗶𝗻𝗲𝘀𝘀 𝗰𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲 𝗼𝗿 𝗽𝗿𝗼𝗯𝗹𝗲𝗺 𝘆𝗼𝘂 𝗻𝗲𝗲𝗱𝗲𝗱 𝘁𝗼 𝘀𝗼𝗹𝘃𝗲
Agriculture has always been one of the core economic sectors in Thailand, with over 40% of Thai workers employed in the industry. However, this industry contributes to only 10% of the economy and is on a decline. Fortunately, Thailand’s government is driving a new initiative by developing new technologies to help transform its agriculture industry.
One of such projects that the government has been supporting is an application called Agri-Map. Agri-map is a collaboration between the Ministry of Agriculture and NSTDA – an initiative to habituate farmers to the environmental changes which affect the country’s agricultural products. The application allows Thai farmers to understand the conditions of their land, knowing what to grow in each specific area instead of only growing a single type of crop for harvest.
Jutarat Khiripet is a researcher in KEA – who has taken on a mission of developing standard metadata for data sharing and integration based on open source and open standards system. The laboratory applies data warehouse and data mining methodologies to develop knowledge elicitation and archive of requirements from government and various industry sectors. In order to project the result into Agri-Map, there is a lot of analytical work to do behind the stage. Prior to Alteryx, Jutarat was simply overwhelmed by the amount of data from their partners – such as the Land Development Department, whose data came as vector files (GIS data) and other data that her team had to find such as the weather forecast API, and opened data sources. As it’s a national project, the data that the team got were immense: more than 100 layers of data needed geospatial process implementation. She calculated and found that in order to do data cleansing and preparation alone, it would take the team of 6 people (an operation specialist, developers, application specialists, and a coordinator) 1 week to finish only 1 layer as there are 1-8 files are included in one layer alone.
𝗗𝗲𝘀𝗰𝗿𝗶𝗯𝗲 𝘆𝗼𝘂𝗿 𝘄𝗼𝗿𝗸𝗶𝗻𝗴 𝘀𝗼𝗹𝘂𝘁𝗶𝗼𝗻
In order to gain insight in the agricultural users of land, NECTEC was interested in leveraging Google Big Query in order to perform a large number of spatial queries. And by using Alteryx Designer, the NECTEC team was able to integrate Python code in order to upload large amounts of data and send queries to Big Query as well as download and integrate the results into a separate database.
There are 3 processes that the NECTEC team uses Alteryx Designer to improve their productivity: –
- To do data cleansing, preparation, and verification for the geo process.
- To help identify layers, area calculation using spatial analysis as well as calculating other statistical information such as cost and benefits, plants, quality of soil, etc.
- Project the refined data onto the Agri-Map Application
Recently, the NECTEC team also adopted Alteryx Server and Alteryx Promote to perform modeling, deploy the data, and share the workflows within the organization. The Alteryx Server is used mainly for data preparation, together with data cleansing. The team is also currently in the process of training the model. As for Alteryx Promote, the team is planning to use it to deploy new models to improve the performance of the data, as there are requests for real-time data through the Agri-map application from any time and anywhere in the country. Currently, the data in the Agri-map application are fixed data – without real-time data of the specific spot, the Agri-map would not be able to provide the objective it set to achieve. The team wants to use Alteryx Promote to improve the real-time ability of data providing and prediction, for example, the weather, the quality of water, or some other variable factors. The Alteryx Designer, Server and Promote are going to be integrated holistically to improve the work of the NECTEC team.
𝗗𝗲𝘀𝗰𝗿𝗶𝗯𝗲 𝘁𝗵𝗲 𝗯𝗲𝗻𝗲𝗳𝗶𝘁𝘀 𝘆𝗼𝘂 𝗵𝗮𝘃𝗲 𝗮𝗰𝗵𝗶𝗲𝘃𝗲𝗱
With Alteryx, Jutarat feels that it helped the NECTEC team be more productive, getting things done faster. Alteryx helps the team to automate processes helping them to reduce the time of data verification and they are able to give their feedback on the data to their partners quicker. The overall processes of working on one layer are completed in 2 days instead of a week. Alteryx also helps them to use the time more effectively since they have more time to do more meaningful work rather than spending it on manual processes.